NLP artificial intelligence method based on financial vertical mapping knowledge domain and interaction system

Abstract

The invention discloses a NLP artificial intelligence method based on a financial vertical mapping knowledge domain and an interaction system. A set of perfect intelligent financial issue interactionsystem is established through association between the established financial vertical mapping knowledge domain and an NLP or natural language processing namely. Different from the passive response of traditional searching and question-answering systems, the method can carry out active rhetorical question, and recommend and excavate problems that users want to learn in a deep level; the flow of themethod includes steps of receiving a user question, carrying out NLP technology, and then analyzing semantic; indexing the financial vertical mapping knowledge domain through the semantic, selecting ascene scale of the question answer, and selecting the most suitable answer and pushing to customers so that questions mentioned by customers can be better understood by combining with the mapping knowledge domain while more accurately indexed.

Description

technical field [0001] The invention relates to the field of artificial intelligence dialogue system and the field of robot language. Background technique [0002] At present, financial managers generally act as shopping guides in the process of purchasing financial products, and customers need continuous communication and companionship during the entire thinking and decision-making process. Unlike traditional product sales, for financial products, purchasing is only the beginning, and the follow-up process requires more communication to play the role of caring and companionship. However, it is obviously impossible to use human services to provide communication around the clock and ensure high-quality and efficient services. The rise of AI intelligence has made it technically possible to provide efficient and intelligent services to customers around the clock. [0003] In the prior art, the technologies that the man-machine dialogue system relies on are mainly divided into...

AI Technical Summary

Problems solved by technology

[0007] 1. The questions and corresponding answers of the traditional method are stored relatively independently, and there is no correlation between each question, so it is impossible to effectively infer user questions through semantics;

[0008] 2. Traditional methods are more focused on solving common user problems and pre-sales problems, and cannot accurately answer most personalized questions after sales;

[0009] 3. The traditional method is based on template matching. The richness of questions is obtained by the generalization of a large number of artificial questions. The same question needs to think of various questions, which takes a lot of time and energy.

Images